Re: pg_basebackup, pg_receivexlog and data durability (was: silent data loss with ext4 / all current versions)

On Thu, Sep 29, 2016 at 8:33 AM, Peter Eisentraut <
peter(dot)eisentraut(at)2ndquadrant(dot)com> wrote:

> On 9/26/16 10:34 PM, Michael Paquier wrote:
> > I thought that as long as the error string is shown to the user, it
> > does not matter much if errno is still saved or not. All the callers
> > of durable_rename() on frontends don't check for strerrno(errno)
> > afterwards. Do you think it matters? Changing that back is trivial.
> > Sorry for not mentioning directly in the thread that I modified that
> > when dropping the last patch set.
>
> Actually, I think the equivalent backend code only does this to save the
> errno across the close call because the elog call comes after the close.
> So it's OK to not do that in the frontend.
>
> With that in mind, I have committed the v3 series now.
>

I'm getting compiler warnings:

file_utils.c: In function 'fsync_pgdata':
file_utils.c:86: warning: passing argument 2 of 'walkdir' from incompatible
pointer type
file_utils.c:36: note: expected 'int (*)(const char *, bool, const char
*)' but argument is of type 'void (*)(const char *, bool, const char *)'
file_utils.c:88: warning: passing argument 2 of 'walkdir' from incompatible
pointer type
file_utils.c:36: note: expected 'int (*)(const char *, bool, const char
*)' but argument is of type 'void (*)(const char *, bool, const char *)'
file_utils.c:89: warning: passing argument 2 of 'walkdir' from incompatible
pointer type
file_utils.c:36: note: expected 'int (*)(const char *, bool, const char
*)' but argument is of type 'void (*)(const char *, bool, const char *)'

Cheers,

Jeff